Designing data warehouses with OO conceptual models

1
Designing data warehouses with OO conceptual
models
?? ???
2
Outline

• Introduction
• OO Conceptual Modeling Approach
• Conclusion

3
Introduction

• The multidimensional model closely parallels how
  data analyzers think and, therefore, helps users
  understand data.
• This approach helps predict what final users want
  to do, thereby facilitating performance
  improvements.

4
Automatic generation of the database schemes

• Initial user requirements.
• System behavior.
• Available data sources.
• Specific issues related.

5

• Multidimensional modeling properties.
• Dimensional fact (DF)
• Multidimensional/entity-relationship (M/ER)
• StarEr model (StarEr )

6
A multi-dimensional model data cube
7
(No Transcript)
8
OO Conceptual Modeling Approach

• Uses a UML class diagram to specify the structure
  of a multidimensional model.
• Structure level
• Clearly separates the structure of a
  multidimensional model specified with a UML class
  diagram into facts and dimensions.
• Facts and dimensions

9
(No Transcript)
10
Classification hierarchies
11
Categorizing dimensions
12
Dynamic level
13
Computer-assisted Software Engineering tool
14
(No Transcript)
15

• Structural
• the structures that form the database schema for
  housing multidimensional data and the underlying
  multidimensional modelalso known as the
  metadatathat provides the models key semantics,
  such as facts, measures, and dimensions.
• Dynamic
• refers to the definition of final user
  requirements and OLAP operations for further
  analyzing data.

16
Informix MetaCube Decision Support System
17
Conclusion

• Integrate commercial OLAP tool facilities within
  GOLD Model Case Tool.
• Data warehouse prototyping.
• Data generation issue.